Concept of model based tampering for improving process performance: an illustrative application to turning process

摘要

This paper presents the concept of a methodology called Model Based Tampering (MBT), that considers the non-linear and stochastic nature of process dynamics, to compensate, in real-time, the effects of process degradation on the performance. The uniqueness of this concept emerges from the ability to combine and utilize the structures of the existing models to fit tractable and robust real-time control models in the form of low-dimensional nonlinear stochastic differential equations (n-SDEs). As an illustrative application of MBT, the minimization of diameter variation due to tool degradation in the turning process is considered. Through a series of simulation runs, the use of MBT was found to reduce diameter variation by 97%. The methodology is applicable to a wide range of manufacturing processes.